點雲結構化為被廣泛研究之課題，現行研究包含從重建點雲物件幾何到運用深度學習模型實現Scan-to-BIM任務。然而，現行研究多針對特定物件，如建物屋頂面或其它接近尺寸之點雲物件，以預設的幾何條件擬定演算策略。當物件具有不同幾何特徵或不同尺度的幾何變化時，現今演算法的適應性及成果的細緻度仍有改善的空間。同時，不論是基於幾何關係或是深度學習演算法之成果，往往需要人工介入以確保物件頂角隅特徵點連線關係之正確與簡潔，以進而產製具有合理特徵連線關係的向量模型。
因此，本研究提出特徵關係向量(Feature Relationship Algorithm, FRA)演算法，由離散點雲中透過研究中提出的掃描球(ScanningBall)建構局部特徵點描述，在考量應用所需的最小空間解析度、資料品質與點密度之參數配置下，可避免忽略同一物件中不同尺度之幾何變化，萃取出可描述物件幾何之頂角隅特徵點(VertexNode)與邊緣特徵點(EdgeNode)，萃取之VertexNode與EdgeNode繼承幾何資訊包含掃描球中心點坐標、特徵點種類、點位編號、局部幾何參數，以及研究研擬混合式評分指標SI(ScanningBall Index)為各萃取特徵點進行穩健度檢核計算。完成特徵點萃取後，FRA演算中共有三大連線關係演算分別為VertexNode to VertexNode(V2V)、VertexNode to EdgeNode (V2E)、EdgeNode to EdgeNode (E2E)，自動重建每一特徵點間之連線關係，進而獲得可描述物件幾何之線框模型，不管是在Scan-to-BIM或是物件點雲模型重建等應用皆具有高度的應用潛力。
研究重於演算法之適應性、自動化與成果細緻度，同時探討並提供在處理不同品質與密度點雲下的參數配置。研究中深入探討掃描球之球半徑、雜訊干擾二者對於演算法之影響，提供相對應之演算參數配置。初步藉由模擬資料BIM-to-Point cloud與實際資料驗證成果顯示，於BIM-to-Point cloud測試資料中萃取出的VertexNode RMSE可達到5公厘之細緻成果。點間距平均為3公分之實際資料，VertexNode之RMSE為4.3公分。FRA演算法可同時偵測出公厘至公尺尺度下物件幾何變化的特徵點，並成功產製出完整的物件線框模型。有鑑於此，本研究成果可望獲得提升在處理具有不同幾何與不同尺度構件之點雲模型重建任務下的自動化及細緻程度，並促進點雲加值應用之效益。

This study proposes a feature relationship algorithm (FRA) to establish wireframe models from object point clouds. To this end, FRA applies a scanning ball to explore local point cloud geometry for determining vertex and edge feature points and further reforms the spatial connections to reconstruct the wireframe model of the object automatically. The scanning ball conveys twofold information to the FRA. The first is feature category, the VertexNode and the EdgeNode, of the scanning ball center along with a hybrid scoring Scanning Index (SI), and the other one is the spatial relationships with neighboring nodes referring to candidates’ pool of feature nodes. FRA purpose three innovative feature-wire description, which are the VertexNode to VertexNode (V2V), VertexNode to EdgeNode (V2E) and EdgeNode to EdgeNode (E2E). These three feature-wire descriptions enhance geometric information exchange between feature points.
Experiments on different variables were conducted to show insights into the effeteness of different point quality for FRA parameter configuration. Also, the level of detail in dealing with an object that contains multi-scale geometry was verified. Finally, validations on data acquired from simulated and actual scans showed promising results, in which the RMSE of the model vertices achieved 3 to 8 mm in BIM-to-point cloud cases, the RMSE of the actual point cloud with point interval of 3cm data case comes to 4.3cm. The FRA framework is expected to improve the automated level of point cloud modeling.

Ali, W., Liu, P., Ying, R., Gong, Z., 2021. A Feature Based Laser SLAM Using Rasterized Images of 3D Point Cloud. IEEE Sensors Journal 21, 24422–24430. https://doi.org/10.1109/JSEN.2021.3113304
Axelsson, P., 1999. Processing of laser scanner data—algorithms and applications. ISPRS Journal of Photogrammetry and Remote Sensing 54, 138–147.
Borrmann, D., Elseberg, J., Lingemann, K., Nüchter, A., 2011. The 3d hough transform for plane detection in point clouds: A review and a new accumulator design. 3D Research 2, 1–13.
Canny, J., 1986. A computational approach to edge detection. IEEE Transactions on pattern analysis and machine intelligence 679–698.
Chuang, T.-Y., Sung, C.-C., 2021. Learning-guided point cloud vectorization for building component modeling. Automation in Construction 132, 103978. https://doi.org/10.1016/j.autcon.2021.103978
Comba, L., Biglia, A., Aimonino, D.R., Gay, P., 2018. Unsupervised detection of vineyards by 3D point-cloud UAV photogrammetry for precision agriculture. Computers and Electronics in Agriculture 155, 84–95.
Cui, Y., Chen, R., Chu, W., Chen, L., Tian, D., Li, Y., Cao, D., 2022. Deep Learning for Image and Point Cloud Fusion in Autonomous Driving: A Review. IEEE Transactions on Intelligent Transportation Systems 23, 722–739. https://doi.org/10.1109/TITS.2020.3023541
Dimitrov, A., Golparvar-Fard, M., 2015. Segmentation of building point cloud models including detailed architectural/structural features and MEP systems. Automation in Construction 51, 32–45.
Grilli, E., Menna, F., Remondino, F., 2017. A review of point clouds segmentation and classification algorithms. The International Archives of Photogrammetry, Remote Sensing and Spatial Information Sciences 42, 339.
Han, H., Han, X., Sun, F., Huang, C., 2015. Point cloud simplification with preserved edge based on normal vector. Optik-International Journal for Light and Electron Optics 126, 2157–2162.
Han, W., Wen, C., Wang, C., Li, X., Li, Q., 2020. Point2Node: Correlation Learning of Dynamic-Node for Point Cloud Feature Modeling. Proceedings of the AAAI Conference on Artificial Intelligence 34, 10925–10932. https://doi.org/10.1609/aaai.v34i07.6725
Huang, Z., Wen, Y., Wang, Z., Ren, J., Jia, K., 2022. Surface Reconstruction from Point Clouds: A Survey and a Benchmark. arXiv preprint arXiv:2205.02413.
 
Illingworth, J., Kittler, J., 1988. A survey of the Hough transform. Computer vision, graphics, and image processing 44, 87–116.
Jolliffe, I.T., 2002. Principal component analysis for special types of data. Springer.
Khemlani, L., 2004. Autodesk Revit: implementation in practice. White paper, Autodesk.
Li, Y., Bu, R., Sun, M., Wu, W., Di, X., Chen, B., 2018. Pointcnn: Convolution on x-transformed points. Advances in neural information processing systems 31.
Li, Y., Olson, E.B., 2010. Extracting general-purpose features from LIDAR data, in: 2010 IEEE International Conference on Robotics and Automation. IEEE, pp. 1388–1393.
Lin, C.-C., Tai, Y.-C., Lee, J.-J., Chen, Y.-S., 2017. A novel point cloud registration using 2D image features. EURASIP Journal on Advances in Signal Processing 2017, 5. https://doi.org/10.1186/s13634-016-0435-y
Liu, W., Wang, C., Chen, S., Bian, X., Lai, B., Shen, X., Cheng, M., Lai, S.-H., Weng, D., Li, J., 2021. Y-Net: Learning Domain Robust Feature Representation for ground camera image and large-scale image-based point cloud registration. Information Sciences 581, 655–677. https://doi.org/10.1016/j.ins.2021.10.022
Lyu, Y., Huang, X., Zhang, Z., 2020. Learning to segment 3d point clouds in 2d image space, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 12255–12264.
Ma, Z., Liu, S., 2018. A review of 3D reconstruction techniques in civil engineering and their applications. Advanced Engineering Informatics 37, 163–174.
Maas, H.-G., Vosselman, G., 1999. Two algorithms for extracting building models from raw laser altimetry data. ISPRS Journal of photogrammetry and remote sensing 54, 153–163.
Meagher, D., 1982. Geometric modeling using octree encoding. Computer Graphics and Image Processing 19, 129–147. https://doi.org/10.1016/0146-664X(82)90104-6
Min, T., Kim, E., Shim, I., 2021. Geometry Guided Network for Point Cloud Registration. IEEE Robotics and Automation Letters 6, 7270–7277. https://doi.org/10.1109/LRA.2021.3097268
Najibi, M., Lai, G., Kundu, A., Lu, Z., Rathod, V., Funkhouser, T., Pantofaru, C., Ross, D., Davis, L.S., Fathi, A., 2020. Dops: Learning to detect 3d objects and predict their 3d shapes, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 11913–11922.
Pang, G., Qiu, R., Huang, J., You, S., Neumann, U., 2015. Automatic 3D Industrial Point Cloud Classification and Modeling. https://doi.org/10.2118/174069-MS
 
Park, J., Cho, Y.K., 2022. Point Cloud Information Modeling: Deep Learning–Based Automated Information Modeling Framework for Point Cloud Data. Journal of Construction Engineering and Management 148, 04021191.
Park, M.K., Lee, S.J., Lee, K.H., 2012. Multi-scale tensor voting for feature extraction from unstructured point clouds. Graphical Models 74, 197–208.
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., 2011. Scikit-learn: Machine learning in Python. the Journal of machine Learning research 12, 2825–2830.
Priestnall, G., Jaafar, J., Duncan, A., 2000. Extracting urban features from LiDAR digital surface models. Computers, Environment and Urban Systems 24, 65–78.
Qi, Charles R., Su, H., Mo, K., Guibas, L.J., 2017. Pointnet: Deep learning on point sets for 3d classification and segmentation, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. pp. 652–660.
Qi, Charles Ruizhongtai, Yi, L., Su, H., Guibas, L.J., 2017. Pointnet++: Deep hierarchical feature learning on point sets in a metric space. Advances in neural information processing systems 30.
Rabbani, T., Van Den Heuvel, F., Vosselmann, G., 2006. Segmentation of point clouds using smoothness constraint. International archives of photogrammetry, remote sensing and spatial information sciences 36, 248–253.
Rao, Y., Fan, B., Wang, Q., Pu, J., Luo, X., Jin, R., 2019. Extreme Feature Regions Detection and Accurate Quality Assessment for Point-Cloud 3D Reconstruction. IEEE Access 7, 37757–37769. https://doi.org/10.1109/ACCESS.2019.2898731
Schnabel, R., Wahl, R., Klein, R., 2007. Efficient RANSAC for point‐cloud shape detection, in: Computer Graphics Forum. Wiley Online Library, pp. 214–226.
Sithole, G., Vosselman, G., 2003. Automatic structure detection in a point-cloud of an urban landscape, in: 2003 2nd GRSS/ISPRS Joint Workshop on Remote Sensing and Data Fusion over Urban Areas. Presented at the 2003 2nd GRSS/ISPRS Joint Workshop on Remote Sensing and Data Fusion over Urban Areas, pp. 67–71. https://doi.org/10.1109/DFUA.2003.1219959
Tang, P., Huber, D., Akinci, B., Lipman, R., Lytle, A., 2010. Automatic reconstruction of as-built building information models from laser-scanned point clouds: A review of related techniques. Automation in construction 19, 829–843.
Van Der Walt, S., Colbert, S.C., Varoquaux, G., 2011. The NumPy array: a structure for efficient numerical computation. Computing in science & engineering 13, 22–30.
Vovk, V., 2013. Kernel ridge regression, in: Empirical Inference. Springer, pp. 105–116.
Wang, C., Hou, S., Wen, C., Gong, Z., Li, Q., Sun, X., Li, J., 2018. Semantic line framework-based indoor building modeling using backpacked laser scanning point cloud. ISPRS journal of photogrammetry and remote sensing 143, 150–166.
Wang, R., Peethambaran, J., Chen, D., 2018. LiDAR Point Clouds to 3-D Urban Models$:$ A Review. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 11, 606–627. https://doi.org/10.1109/JSTARS.2017.2781132
Wu, B., Ma, J., Chen, G., An, P., 2021. Feature interactive representation for point cloud registration, in: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 5530–5539.
Yoo, H.W., Kim, W.H., Park, J.W., Lee, W.H., Chung, M.J., 2013. Real-time plane detection based on depth map from kinect, in: IEEE ISR 2013. IEEE, pp. 1–4.
You, R.-J., Lin, B.-C., 2011. Building feature extraction from airborne lidar data based on tensor voting algorithm. Photogrammetric Engineering & Remote Sensing 77, 1221–1231.
Zhang, M., You, H., Kadam, P., Liu, S., Kuo, C.-C.J., 2020. Pointhop: An explainable machine learning method for point cloud classification. IEEE Transactions on Multimedia 22, 1744–1755.
Zhang, Z., Dai, Y., Sun, J., 2020. Deep learning based point cloud registration: an overview. Virtual Reality & Intelligent Hardware 2, 222–246. https://doi.org/10.1016/j.vrih.2020.05.002
Zhou, Q.-Y., Park, J., Koltun, V., 2018. Open3D: A modern library for 3D data processing. arXiv preprint arXiv:1801.09847.
Zhou, Y., Tuzel, O., 2018. Voxelnet: End-to-end learning for point cloud based 3d object detection, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. pp. 4490–4499.
張中豪, 2013. 自適性張量分析應用於光達點雲特徵萃取. 臺灣大學土木工程學研究所學位論文. 國立臺灣大學. https://doi.org/10.6342/NTU.2013.02134
王名玉, 2010. 多尺度張量應用於 3D 點雲套合之研究.
王淼, 湯凱佩, 曾義星, 2005. 光達資料八分樹結構化於平面特徵萃取. Journal of Photogrammetry and Remote Sensing 10, 59–70.
蔡展榮, 李姝儀, 2006. 萃取物表角面特徵點供多測站地面光達點雲面之連結. Journal of Photogrammetry and Remote Sensing 11, 339–359.
 
賴志凱, 2004. 地面雷射掃瞄儀的精度分析與檢定, 國立成功大學測量及空間資訊學系碩士論文. Lai, CK.
黃國彥, 2005. 結合三維網格及迭代式霍夫轉換進行光達點雲資料結構化處理.